Main Effects
# FLOWERING DURATION
#BB_final_full$log_duration <- log(BB_final_full$fl_duration+1)
#BB_duration_Model = lmer(log_duration ~ watered + (1|Block) + (1|Recipient), data = BB_final_full, REML = T)
#*Best* simple linear model -log and sqrt both work?? sqrt+1 results in nice residuals, better adj R2
PL_Fdur_lm0 <- lm(fl_duration ~ Treatment, data = BB_allplants)
summary(PL_Fdur_lm0)
##
## Call:
## lm(formula = fl_duration ~ Treatment, data = BB_allplants)
##
## Residuals:
## Min 1Q Median 3Q Max
## -17.702 -11.685 -3.702 8.315 50.315
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 15.6939 0.7443 21.086 < 2e-16 ***
## Treatment1 -3.0086 0.7443 -4.042 6.66e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 12.89 on 316 degrees of freedom
## (522 observations deleted due to missingness)
## Multiple R-squared: 0.04917, Adjusted R-squared: 0.04616
## F-statistic: 16.34 on 1 and 316 DF, p-value: 6.656e-05
plot(PL_Fdur_lm0)
hist(PL_Fdur_lm0$residuals)
##UNSURE which transformation is best!
#John's model equivalent
PL_Fdur_lm1 <- lmer(log(fl_duration+1) ~ Treatment + (1|Recipient) + (1|Block), data = BB_allplants, REML = T)
summary(PL_Fdur_lm1)
## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: log(fl_duration + 1) ~ Treatment + (1 | Recipient) + (1 | Block)
## Data: BB_allplants
##
## REML criterion at convergence: 939.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.8256 -1.1051 0.1926 0.7749 1.7273
##
## Random effects:
## Groups Name Variance Std.Dev.
## Recipient (Intercept) 0.00535 0.07314
## Block (Intercept) 0.03206 0.17906
## Residual 1.09119 1.04460
## Number of obs: 318, groups: Recipient, 21; Block, 4
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 2.3611 0.1210 1.7219 19.509 0.00506 **
## Treatment1 -0.1916 0.1200 1.6550 -1.596 0.27616
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## Treatment1 -0.425
r.squaredGLMM(PL_Fdur_lm1)
## R2m R2c
## [1,] 0.02983624 0.06199636
#ranef(PL_Fdur_lm1)
#Better model
PL_Fdur_lm2 <- lmer(fl_duration ~ Treatment + (1|Location), data = BB_allplants)
summary(PL_Fdur_lm2)
## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: fl_duration ~ Treatment + (1 | Location)
## Data: BB_allplants
##
## REML criterion at convergence: 2514.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.8393 -0.7100 -0.2060 0.5261 3.8505
##
## Random effects:
## Groups Name Variance Std.Dev.
## Location (Intercept) 22.68 4.762
## Residual 145.45 12.060
## Number of obs: 318, groups: Location, 65
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 15.177 1.046 37.075 14.505 <2e-16 ***
## Treatment1 -2.746 1.046 37.075 -2.625 0.0125 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## Treatment1 -0.419
r.squaredGLMM(PL_Fdur_lm2)
## R2m R2c
## [1,] 0.0407051 0.1701059
ranef(PL_Fdur_lm2)
## $Location
## (Intercept)
## 2 4.13165864
## 4 -0.97051062
## 5 -0.77478678
## 8 4.11188563
## 9 2.09347746
## 10 -2.00422106
## 15 1.96522054
## 20 3.41674719
## 26 -0.05709577
## 30 7.11907454
## 34 3.92911718
## 39 -1.83710687
## 40 -5.50810032
## 44 -0.86750360
## 46 -0.22134211
## 47 -1.90010555
## 50 -1.96411853
## 52 3.12987926
## 53 -1.71603605
## 59 0.19441220
## 60 -0.16278252
## 62 2.70248439
## 64 0.41058212
## 67 1.15587079
## 70 -0.21959796
## 73 -2.00422106
## 76 -1.29106948
## 77 -3.64305102
## 79 -1.17221088
## 80.5 5.25845574
## 83 -0.66855477
## 84 -3.64305102
## 85 2.62463751
## 86 4.92401545
## 87 -2.12307965
## 88 -3.64305102
## 90 -1.98736195
## 99 4.65186034
## 100 -4.01252475
## 107 -2.53611402
## 109 -1.54196172
## 110 -1.60037824
## 112 -3.91406492
## 113 -2.71737264
## 114 -3.81475972
## 115 -3.64305102
## 119 -1.54196172
## 120 5.18062243
## 122 0.79470411
## 125 0.16160218
## 129 -1.12590582
## 130 0.94673670
## 132 -1.54196172
## 134 2.77228019
## 136 -3.00565897
## 140 -4.67621585
## 141 -2.25130887
## 143 6.31684869
## 144 0.67776245
## 146 3.14056753
## 147 3.17924518
## 150 2.28780908
## 151 -0.34020071
## 155 0.02646014
## 160 -0.66161838
##
## with conditional variances for "Location"
plot(PL_Fdur_lm2)
#Log transofrmation seems more appropriate here...? Better R2c
#Location results in singular model but block does not
##FIRST FLOWERING DATE
#BB_FFD_Model = lmer(FFD_DOY ~ watered + (1|Block) + (1|Recipient), data = BB_final_full, REML = T)
#best lm
PL_FFD_lm0 <- lm(FFD_z ~ Treatment, data = BB_allplants)
summary(PL_FFD_lm0)
##
## Call:
## lm(formula = FFD_z ~ Treatment, data = BB_allplants)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.2915 -0.5976 0.0800 0.7142 4.2005
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.006587 0.057821 -0.114 0.909
## Treatment1 0.027563 0.057821 0.477 0.634
##
## Residual standard error: 1.001 on 316 degrees of freedom
## (522 observations deleted due to missingness)
## Multiple R-squared: 0.0007186, Adjusted R-squared: -0.002444
## F-statistic: 0.2272 on 1 and 316 DF, p-value: 0.6339
plot(PL_FFD_lm0)
hist(PL_FFD_lm0$residuals)
#John's model equivalent
PL_FFD_lm1 <- lmer(yday(FFD) ~ Treatment + (1|Recipient) + (1|Block), data = BB_allplants)
## boundary (singular) fit: see help('isSingular')
summary(PL_FFD_lm1)
## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: yday(FFD) ~ Treatment + (1 | Recipient) + (1 | Block)
## Data: BB_allplants
##
## REML criterion at convergence: 2452.2
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.6909 -0.6384 -0.0498 0.4373 4.3709
##
## Random effects:
## Groups Name Variance Std.Dev.
## Recipient (Intercept) 18.76 4.332
## Block (Intercept) 0.00 0.000
## Residual 122.89 11.086
## Number of obs: 318, groups: Recipient, 21; Block, 4
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 83.32860 1.14944 19.49311 72.495 <2e-16 ***
## Treatment1 -0.06952 0.66507 312.12445 -0.105 0.917
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## Treatment1 -0.139
## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')
r.squaredGLMM(PL_FFD_lm1)
## R2m R2c
## [1,] 3.226998e-05 0.1324817
ranef(PL_FFD_lm1) #Model is singular when using Block instead of Location
## $Recipient
## (Intercept)
## BB_017 -6.3388025
## BB_020 -2.0811881
## BB_024 -3.7074741
## BB_032 6.1636658
## BB_049 -1.9748934
## BB_051 -2.5461178
## BB_059 2.5716485
## BB_061 4.0687139
## BB_077 0.1478030
## BB_079 0.1911948
## BB_085 -5.1821042
## BB_102 -2.7303583
## BB_105 -1.1700497
## BB_118 -1.4604022
## BB_137 -0.6482969
## BB_143 1.0148859
## BB_152 -1.7945927
## BB_161 7.4568323
## BB_164 3.8776317
## BB_185 0.2492218
## BB_187 3.8926822
##
## $Block
## (Intercept)
## 1 0
## 2 0
## 3 0
## W 0
##
## with conditional variances for "Recipient" "Block"
#Better model - remove Recipient? Still NS, same results
PL_FFD_lm2 <- lmer(yday(FFD) ~ Treatment + (1|Recipient) + (1|Location), data = BB_allplants)
summary(PL_FFD_lm2)
## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: yday(FFD) ~ Treatment + (1 | Recipient) + (1 | Location)
## Data: BB_allplants
##
## REML criterion at convergence: 2415.5
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.0326 -0.6276 -0.1231 0.5090 4.0075
##
## Random effects:
## Groups Name Variance Std.Dev.
## Location (Intercept) 51.645 7.186
## Recipient (Intercept) 5.187 2.278
## Residual 90.963 9.537
## Number of obs: 318, groups: Location, 65; Recipient, 21
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 84.678 1.355 33.416 62.509 <2e-16 ***
## Treatment1 -0.848 1.262 38.615 -0.672 0.506
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## Treatment1 -0.462
r.squaredGLMM(PL_FFD_lm2)
## R2m R2c
## [1,] 0.004580967 0.3873525
ranef(PL_FFD_lm2)
## $Location
## (Intercept)
## 2 -3.83327034
## 4 -1.23138885
## 5 1.59919320
## 8 -0.16702018
## 9 -1.56117894
## 10 3.60982642
## 15 9.64240525
## 20 -0.02529969
## 26 2.94145706
## 30 1.59661731
## 34 -3.37831371
## 39 2.01303340
## 40 7.33601403
## 44 2.58571417
## 46 -7.03118180
## 47 -3.39617312
## 50 -4.90643401
## 52 -1.59074347
## 53 2.85402807
## 59 -2.97387544
## 60 -5.48498928
## 62 -5.00954654
## 64 -3.76273004
## 67 4.63568332
## 70 13.49842018
## 73 4.66683791
## 76 1.44782698
## 77 9.03998802
## 79 2.05330323
## 80.5 2.28133220
## 83 3.88440216
## 84 5.18571737
## 85 2.33744616
## 86 -12.84165458
## 87 1.98596278
## 88 0.32905302
## 90 2.54058164
## 99 -11.66449390
## 100 -0.02512620
## 107 -2.30939702
## 109 3.17845634
## 110 4.17762700
## 112 10.60866185
## 113 3.45661703
## 114 9.96120712
## 115 -1.52030617
## 119 12.36840662
## 120 -10.68870378
## 122 -6.15137793
## 125 2.39613114
## 129 -4.69689475
## 130 -6.53464697
## 132 -4.33013498
## 134 -11.08874917
## 136 3.96472176
## 140 2.54972529
## 141 2.70504949
## 143 -4.49234725
## 144 -6.16268930
## 146 -8.11111709
## 147 -2.45162161
## 150 1.99901305
## 151 -0.75482712
## 155 1.08520309
## 160 -8.33943043
##
## $Recipient
## (Intercept)
## BB_017 -2.35225741
## BB_020 -0.36810552
## BB_024 -0.60675140
## BB_032 0.82255847
## BB_049 -1.27021717
## BB_051 -1.41044180
## BB_059 1.10693422
## BB_061 1.54421471
## BB_077 0.91620646
## BB_079 -0.87312513
## BB_085 -1.15279262
## BB_102 -0.66164285
## BB_105 0.30841923
## BB_118 -1.83073717
## BB_137 0.43506374
## BB_143 1.01685156
## BB_152 1.02999964
## BB_161 1.94711431
## BB_164 1.36937805
## BB_185 0.08963305
## BB_187 -0.06030239
##
## with conditional variances for "Location" "Recipient"
# COROLLA DIAMETER
#BB_diameter_Model = lmer(corolla_diam_mm ~ watered + (1|Block) + (1|Recipient), data = BB_final, REML = T)
#best lm
PL_Cd_lm0 <- lm(corolla_diam_mm ~ Treatment, data = BB_Fplants)
summary(PL_Cd_lm0)
##
## Call:
## lm(formula = corolla_diam_mm ~ Treatment, data = BB_Fplants)
##
## Residuals:
## Min 1Q Median 3Q Max
## -10.3677 -3.3574 0.1757 3.3192 12.5733
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 23.0273 0.4186 55.011 < 2e-16 ***
## Treatment1 -1.2147 0.4186 -2.902 0.00434 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.687 on 134 degrees of freedom
## (89 observations deleted due to missingness)
## Multiple R-squared: 0.05912, Adjusted R-squared: 0.0521
## F-statistic: 8.42 on 1 and 134 DF, p-value: 0.004339
plot(PL_Cd_lm0)
hist(PL_Cd_lm0$residuals)
#John's equivalent model
PL_Cdiam_lm1 <- lmer(corolla_diam_mm ~ Treatment + (1|Block) + (1|Recipient), data = BB_Fplants)
summary(PL_Cdiam_lm1)
## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: corolla_diam_mm ~ Treatment + (1 | Block) + (1 | Recipient)
## Data: BB_Fplants
##
## REML criterion at convergence: 797.9
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.53414 -0.67403 0.05407 0.69534 2.22526
##
## Random effects:
## Groups Name Variance Std.Dev.
## Recipient (Intercept) 2.433 1.560
## Block (Intercept) 2.424 1.557
## Residual 18.779 4.333
## Number of obs: 136, groups: Recipient, 21; Block, 4
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 22.9315 1.0408 2.1722 22.032 0.00135 **
## Treatment1 -1.3950 0.9835 1.7579 -1.418 0.30744
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## Treatment1 -0.437
r.squaredGLMM(PL_Cdiam_lm1)
## R2m R2c
## [1,] 0.07103367 0.2619411
#ranef(PL_Cdiam_lm1) #Both Location or Block result in singular model...
#Block explains variation in corolla diameter better than treatment or Recipient!
#Better model? Same results
PL_Cdiam_lm2 <- lmer(corolla_diam_mm ~ Treatment + (1|Location) + (1|Recipient), data = BB_Fplants)
summary(PL_Cdiam_lm2)
## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: corolla_diam_mm ~ Treatment + (1 | Location) + (1 | Recipient)
## Data: BB_Fplants
##
## REML criterion at convergence: 801.2
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.28525 -0.62951 0.03972 0.63856 2.37703
##
## Random effects:
## Groups Name Variance Std.Dev.
## Location (Intercept) 1.739 1.319
## Recipient (Intercept) 1.832 1.353
## Residual 18.659 4.320
## Number of obs: 136, groups: Location, 60; Recipient, 21
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 22.9616 0.5439 19.3398 42.217 < 2e-16 ***
## Treatment1 -1.3350 0.4548 23.9872 -2.935 0.00723 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## Treatment1 -0.304
r.squaredGLMM(PL_Cdiam_lm2)
## R2m R2c
## [1,] 0.06929819 0.2188034
#ranef(PL_Cdiam_lm2)
#ABOVEGROUND BIOMASS
#BB_final$log_biomass <- log(BB_final$AG_biomass_mg)
#BB_biomass_Model = lmer(log_biomass ~ watered + (1|Block) + (1|Recipient), data = BB_final, REML = T)
#Best lm
PL_AG_lm0 <- lm(AG_biomass_mg ~ Treatment, data = BB_Fplants)
summary(PL_AG_lm0)
##
## Call:
## lm(formula = AG_biomass_mg ~ Treatment, data = BB_Fplants)
##
## Residuals:
## Min 1Q Median 3Q Max
## -24.288 -12.594 -6.110 7.937 70.382
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 21.784 1.995 10.916 <2e-16 ***
## Treatment1 -3.784 1.995 -1.896 0.061 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 19.36 on 93 degrees of freedom
## (130 observations deleted due to missingness)
## Multiple R-squared: 0.03722, Adjusted R-squared: 0.02687
## F-statistic: 3.596 on 1 and 93 DF, p-value: 0.06103
plot(PL_AG_lm0)
hist(PL_AG_lm0$residuals)
#Transformation unnecessary? Better R2, marginally NS
#John Equivalent Model
PL_AGbio_lm1 <- lmer(log(AG_biomass_mg) ~ Treatment + (1|Location) + (1|Recipient), data = BB_Fplants)
## boundary (singular) fit: see help('isSingular')
summary(PL_AGbio_lm1)
## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: log(AG_biomass_mg) ~ Treatment + (1 | Location) + (1 | Recipient)
## Data: BB_Fplants
##
## REML criterion at convergence: 248.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.47160 -0.62149 0.06617 0.59026 1.92437
##
## Random effects:
## Groups Name Variance Std.Dev.
## Location (Intercept) 0.1508 0.3884
## Recipient (Intercept) 0.0000 0.0000
## Residual 0.6431 0.8019
## Number of obs: 95, groups: Location, 47; Recipient, 21
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 2.67968 0.10738 29.48622 24.955 <2e-16 ***
## Treatment1 -0.06436 0.10738 29.48622 -0.599 0.554
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## Treatment1 -0.244
## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')
r.squaredGLMM(PL_Cdiam_lm1)
## R2m R2c
## [1,] 0.07103367 0.2619411
ranef(PL_Cdiam_lm1) #Both Location or Block result in singular model
## $Recipient
## (Intercept)
## BB_017 -0.18022322
## BB_020 0.01952809
## BB_024 -0.25812589
## BB_032 0.22634043
## BB_049 -0.34396967
## BB_051 1.24023209
## BB_059 0.13983849
## BB_061 -1.97150244
## BB_077 0.41111741
## BB_079 1.35547499
## BB_085 0.64592842
## BB_102 1.29339560
## BB_105 -0.35375886
## BB_118 0.24879740
## BB_137 0.58507656
## BB_143 -0.44847727
## BB_152 -1.18241408
## BB_161 2.19212642
## BB_164 -0.91160214
## BB_185 -1.59430199
## BB_187 -1.11348032
##
## $Block
## (Intercept)
## 1 1.580073e+00
## 2 -8.658975e-01
## 3 -7.141753e-01
## W -8.090670e-14
##
## with conditional variances for "Recipient" "Block"
#Better model? #similar results as lm
PL_AGbio_lm2 <- lmer(AG_biomass_mg ~ Treatment + (1|Location), data = BB_Fplants)
summary(PL_AGbio_lm2)
## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: AG_biomass_mg ~ Treatment + (1 | Location)
## Data: BB_Fplants
##
## REML criterion at convergence: 824
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.1574 -0.6053 -0.2880 0.2975 3.5835
##
## Random effects:
## Groups Name Variance Std.Dev.
## Location (Intercept) 15.99 3.999
## Residual 359.46 18.960
## Number of obs: 95, groups: Location, 47
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 21.724 2.083 30.710 10.428 1.3e-11 ***
## Treatment1 -3.679 2.083 30.710 -1.766 0.0873 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## Treatment1 -0.138
r.squaredGLMM(PL_Cdiam_lm2)
## R2m R2c
## [1,] 0.06929819 0.2188034
ranef(PL_Cdiam_lm2)
## $Location
## (Intercept)
## 2 1.19838277
## 4 0.79035602
## 5 0.02896805
## 8 0.80353050
## 9 0.30937637
## 10 -0.29820937
## 20 0.17022780
## 26 -0.31938080
## 30 0.50341349
## 34 0.82215204
## 39 0.69464557
## 40 -0.05286850
## 44 0.15392778
## 46 0.00597208
## 47 -0.64341118
## 50 0.43066476
## 52 0.54585612
## 53 0.23698897
## 59 -0.27438370
## 60 0.16921935
## 62 -0.64769414
## 64 -0.02947460
## 67 -0.09658465
## 70 -0.81424854
## 73 -0.49651820
## 76 0.44637536
## 77 0.21478981
## 79 -0.49954694
## 80.5 0.58423475
## 83 -0.63099051
## 84 -0.62240612
## 85 0.82187718
## 86 0.06736266
## 87 -0.06240165
## 88 -0.25663933
## 90 -0.56308617
## 100 -0.52678934
## 110 0.24278958
## 112 -0.16651833
## 113 -0.47823729
## 114 -0.22244929
## 115 -0.15501731
## 120 -1.07043748
## 122 -1.27342225
## 125 -0.01636197
## 129 0.08818279
## 130 -0.33855796
## 132 -0.29466771
## 134 -0.14512267
## 136 -0.66177357
## 140 0.40058366
## 141 -0.21796364
## 143 0.40553462
## 144 0.12760664
## 146 0.25228767
## 147 -0.25242739
## 150 0.38362901
## 151 0.24724707
## 155 0.32995477
## 160 0.65145338
##
## $Recipient
## (Intercept)
## BB_017 -0.10687380
## BB_020 0.13401863
## BB_024 -0.04867275
## BB_032 0.05648702
## BB_049 -0.25709783
## BB_051 0.99093569
## BB_059 -0.10039915
## BB_061 -1.61074333
## BB_077 0.16336372
## BB_079 0.80810748
## BB_085 0.55439625
## BB_102 0.95935836
## BB_105 -0.26815302
## BB_118 0.46193623
## BB_137 0.57518627
## BB_143 -0.42551305
## BB_152 -0.99683652
## BB_161 1.70122406
## BB_164 -0.74492948
## BB_185 -0.94752508
## BB_187 -0.89826970
##
## with conditional variances for "Location" "Recipient"
plot(PL_AGbio_lm2)
# FECUNDITY
#BB_final2$log_seedprod <- log(BB_final2$total_est_seed_production)
#BB_seed_Model = lmer(log_seedprod ~ watered + (1|Block) + (1|Recipient), data = BB_final2, REML = T)
#lm
#filtered to individuals that produced seed
PL_fec_lm0 <- lm(log(total_est_seed_production) ~ Treatment, data = filter(BB_Fplants, surv_to_seedprod == 1))
summary(PL_fec_lm0)
##
## Call:
## lm(formula = log(total_est_seed_production) ~ Treatment, data = filter(BB_Fplants,
## surv_to_seedprod == 1))
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.8929 -0.5066 -0.0693 0.5639 3.5329
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.6742 0.1138 14.713 <2e-16 ***
## Treatment1 -0.2186 0.1138 -1.921 0.0583 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.024 on 79 degrees of freedom
## Multiple R-squared: 0.04464, Adjusted R-squared: 0.03254
## F-statistic: 3.691 on 1 and 79 DF, p-value: 0.05831
plot(PL_fec_lm0)
hist(PL_fec_lm0$residuals)
#John's equivalent model
#Excludes plants that didn't survive to produce seeds!!
PL_fec_lm1 <- lmer(log(total_est_seed_production) ~ Treatment + (1|Recipient) + (1|Block), data = filter(BB_Fplants, surv_to_seedprod == 1))
summary(PL_fec_lm1)
## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: log(total_est_seed_production) ~ Treatment + (1 | Recipient) +
## (1 | Block)
## Data: filter(BB_Fplants, surv_to_seedprod == 1)
##
## REML criterion at convergence: 232.5
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.9203 -0.6066 -0.0445 0.6297 3.4282
##
## Random effects:
## Groups Name Variance Std.Dev.
## Recipient (Intercept) 0.187807 0.43337
## Block (Intercept) 0.009578 0.09787
## Residual 0.858017 0.92629
## Number of obs: 81, groups: Recipient, 20; Block, 4
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 1.6233 0.1560 1.1551 10.404 0.044 *
## Treatment1 -0.2174 0.1250 0.4251 -1.739 0.511
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## Treatment1 -0.072
r.squaredGLMM(PL_fec_lm1)
## R2m R2c
## [1,] 0.04336976 0.2222824
#! N = 80 individuals! compared to 225
#Better model?
PL_fec_lm2 <- lmer(log(total_est_seed_production) ~ Treatment + (1|Recipient) + (1|Location), data = filter(BB_Fplants, surv_to_seedprod == 1))
summary(PL_fec_lm2)
## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: log(total_est_seed_production) ~ Treatment + (1 | Recipient) +
## (1 | Location)
## Data: filter(BB_Fplants, surv_to_seedprod == 1)
##
## REML criterion at convergence: 231.6
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.8553 -0.5988 0.0286 0.5352 3.4593
##
## Random effects:
## Groups Name Variance Std.Dev.
## Location (Intercept) 0.1153 0.3396
## Recipient (Intercept) 0.1747 0.4180
## Residual 0.7719 0.8786
## Number of obs: 81, groups: Location, 35; Recipient, 20
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 1.6158 0.1543 19.4928 10.474 1.9e-09 ***
## Treatment1 -0.2135 0.1259 20.3054 -1.696 0.105
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## Treatment1 -0.090
r.squaredGLMM(PL_fec_lm2)
## R2m R2c
## [1,] 0.04162432 0.3033846
##Use GLM?
## SEED COUNT - not analyzed
#lm
GxE Effects
Based on John’s models - (minor improvements made)
#FFD
GxE_FFD_lm0 <- lm(yday(FFD) ~ Recipient*Treatment + Location, data = filter(BB_allplants, Recipient != "BB_161"))
summary(GxE_FFD_lm0)
##
## Call:
## lm(formula = yday(FFD) ~ Recipient * Treatment + Location, data = filter(BB_allplants,
## Recipient != "BB_161"))
##
## Residuals:
## Min 1Q Median 3Q Max
## -28.032 -6.467 -0.690 4.817 42.180
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 87.22535 1.50938 57.789 < 2e-16 ***
## Recipient1 -9.09077 2.37896 -3.821 0.000165 ***
## Recipient2 -2.32999 2.40914 -0.967 0.334347
## Recipient3 -4.77895 2.92224 -1.635 0.103150
## Recipient4 5.77052 3.16597 1.823 0.069472 .
## Recipient5 -3.42764 2.34590 -1.461 0.145158
## Recipient6 -3.20894 2.93239 -1.094 0.274807
## Recipient7 5.62413 2.83284 1.985 0.048131 *
## Recipient8 8.78497 3.39314 2.589 0.010152 *
## Recipient9 3.98402 3.19856 1.246 0.214017
## Recipient10 0.97625 3.43385 0.284 0.776400
## Recipient11 -5.11864 2.91841 -1.754 0.080593 .
## Recipient12 -2.99649 2.27542 -1.317 0.189003
## Recipient13 -0.77401 3.33822 -0.232 0.816822
## Recipient14 -2.27691 2.39086 -0.952 0.341786
## Recipient15 -1.10816 2.86246 -0.387 0.698963
## Recipient16 0.12453 2.79350 0.045 0.964476
## Recipient17 -2.03517 2.89688 -0.703 0.482956
## Recipient18 7.67621 3.81514 2.012 0.045220 *
## Recipient19 -1.02998 2.89744 -0.355 0.722511
## Treatment1 -1.30049 0.69598 -1.869 0.062780 .
## Location -0.03942 0.01472 -2.678 0.007860 **
## Recipient1:Treatment1 5.02344 2.37557 2.115 0.035389 *
## Recipient2:Treatment1 1.61660 2.41552 0.669 0.503911
## Recipient3:Treatment1 4.78964 2.92676 1.636 0.102914
## Recipient4:Treatment1 8.86384 3.18526 2.783 0.005774 **
## Recipient5:Treatment1 2.38927 2.34919 1.017 0.310046
## Recipient6:Treatment1 -0.45880 2.93977 -0.156 0.876099
## Recipient7:Treatment1 -3.61472 2.81730 -1.283 0.200589
## Recipient8:Treatment1 -4.83286 3.39361 -1.424 0.155584
## Recipient9:Treatment1 -12.88624 3.20604 -4.019 7.6e-05 ***
## Recipient10:Treatment1 -0.33749 3.41539 -0.099 0.921359
## Recipient11:Treatment1 -4.64036 2.90738 -1.596 0.111659
## Recipient12:Treatment1 -0.82911 2.26674 -0.366 0.714824
## Recipient13:Treatment1 0.29426 3.31801 0.089 0.929399
## Recipient14:Treatment1 5.18895 2.39016 2.171 0.030815 *
## Recipient15:Treatment1 0.31040 2.86246 0.108 0.913730
## Recipient16:Treatment1 -0.43000 2.73910 -0.157 0.875373
## Recipient17:Treatment1 -0.49570 2.86288 -0.173 0.862666
## Recipient18:Treatment1 -2.34550 3.81842 -0.614 0.539567
## Recipient19:Treatment1 0.12997 2.88240 0.045 0.964069
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 10.52 on 267 degrees of freedom
## (492 observations deleted due to missingness)
## Multiple R-squared: 0.2817, Adjusted R-squared: 0.1741
## F-statistic: 2.618 on 40 and 267 DF, p-value: 2.738e-06
plot(GxE_FFD_lm0)
hist(GxE_FFD_lm0$residuals)
Anova(GxE_FFD_lm0, type = 3)
## Anova Table (Type III tests)
##
## Response: yday(FFD)
## Sum Sq Df F value Pr(>F)
## (Intercept) 369363 1 3339.5358 < 2.2e-16 ***
## Recipient 5152 19 2.4517 0.0008646 ***
## Treatment 386 1 3.4915 0.0627795 .
## Location 793 1 7.1729 0.0078602 **
## Recipient:Treatment 4676 19 2.2252 0.0028349 **
## Residuals 29531 267
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
GxE_FFD_lm_rs <- lm(yday(FFD) ~ Recipient*Treatment + Location, data = BB_allplants)
#Flowering Duration
GxE_Fdur_lm0 <- lm(fl_duration ~ Recipient*Treatment + Location, data = filter(BB_allplants, Recipient != "BB_161"))
summary(GxE_Fdur_lm0)
##
## Call:
## lm(formula = fl_duration ~ Recipient * Treatment + Location,
## data = filter(BB_allplants, Recipient != "BB_161"))
##
## Residuals:
## Min 1Q Median 3Q Max
## -31.264 -8.390 -1.288 6.941 42.044
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 17.99930 1.82339 9.871 < 2e-16 ***
## Recipient1 10.46511 2.87387 3.641 0.000326 ***
## Recipient2 1.22298 2.91033 0.420 0.674662
## Recipient3 0.17556 3.53018 0.050 0.960374
## Recipient4 -4.49749 3.82461 -1.176 0.240669
## Recipient5 -3.39107 2.83393 -1.197 0.232525
## Recipient6 0.56485 3.54244 0.159 0.873433
## Recipient7 4.80898 3.42218 1.405 0.161113
## Recipient8 -2.14524 4.09905 -0.523 0.601163
## Recipient9 -2.13629 3.86399 -0.553 0.580815
## Recipient10 5.63734 4.14822 1.359 0.175300
## Recipient11 -2.42216 3.52555 -0.687 0.492660
## Recipient12 4.04712 2.74880 1.472 0.142111
## Recipient13 0.44001 4.03270 0.109 0.913197
## Recipient14 -1.41923 2.88825 -0.491 0.623560
## Recipient15 -1.03614 3.45796 -0.300 0.764686
## Recipient16 -6.53468 3.37465 -1.936 0.053874 .
## Recipient17 -0.57548 3.49954 -0.164 0.869506
## Recipient18 -1.31903 4.60883 -0.286 0.774950
## Recipient19 2.78854 3.50022 0.797 0.426348
## Treatment1 -2.98569 0.84078 -3.551 0.000453 ***
## Location -0.03205 0.01778 -1.803 0.072589 .
## Recipient1:Treatment1 -4.66012 2.86978 -1.624 0.105585
## Recipient2:Treatment1 -5.48812 2.91804 -1.881 0.061093 .
## Recipient3:Treatment1 -6.56504 3.53564 -1.857 0.064438 .
## Recipient4:Treatment1 -5.22394 3.84792 -1.358 0.175736
## Recipient5:Treatment1 1.47452 2.83791 0.520 0.603788
## Recipient6:Treatment1 2.46372 3.55136 0.694 0.488449
## Recipient7:Treatment1 -1.96776 3.40340 -0.578 0.563633
## Recipient8:Treatment1 4.07510 4.09961 0.994 0.321112
## Recipient9:Treatment1 5.13426 3.87301 1.326 0.186089
## Recipient10:Treatment1 -1.80030 4.12592 -0.436 0.662943
## Recipient11:Treatment1 8.42851 3.51223 2.400 0.017091 *
## Recipient12:Treatment1 1.92567 2.73831 0.703 0.482522
## Recipient13:Treatment1 -1.40791 4.00828 -0.351 0.725677
## Recipient14:Treatment1 -4.13519 2.88741 -1.432 0.153271
## Recipient15:Treatment1 2.00945 3.45796 0.581 0.561657
## Recipient16:Treatment1 2.67104 3.30893 0.807 0.420257
## Recipient17:Treatment1 1.77178 3.45847 0.512 0.608863
## Recipient18:Treatment1 2.78163 4.61280 0.603 0.547004
## Recipient19:Treatment1 -1.76890 3.48205 -0.508 0.611869
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 12.7 on 267 degrees of freedom
## (492 observations deleted due to missingness)
## Multiple R-squared: 0.1995, Adjusted R-squared: 0.07955
## F-statistic: 1.663 on 40 and 267 DF, p-value: 0.01042
plot(GxE_Fdur_lm0)
hist(GxE_Fdur_lm0$residuals)
#Significant GxE effect when log-transformed but not sqrt()
Anova(GxE_Fdur_lm0, type = 3)
## Anova Table (Type III tests)
##
## Response: fl_duration
## Sum Sq Df F value Pr(>F)
## (Intercept) 15728 1 97.4430 < 2.2e-16 ***
## Recipient 4393 19 1.4325 0.1110745
## Treatment 2035 1 12.6104 0.0004533 ***
## Location 524 1 3.2491 0.0725889 .
## Recipient:Treatment 3983 19 1.2988 0.1833232
## Residuals 43096 267
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#Corolla Diam
GxE_Cd_lm0 <- lm(corolla_diam_mm ~ Recipient*Treatment + Location, data = BB_Fplants)
summary(GxE_Cd_lm0)
##
## Call:
## lm(formula = corolla_diam_mm ~ Recipient * Treatment + Location,
## data = BB_Fplants)
##
## Residuals:
## Min 1Q Median 3Q Max
## -9.329 -2.504 0.000 2.423 9.409
##
## Coefficients: (1 not defined because of singularities)
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 24.831461 2.166052 11.464 <2e-16 ***
## Recipient1 -0.421356 2.316598 -0.182 0.8561
## Recipient2 0.047907 2.558282 0.019 0.9851
## Recipient3 0.276518 3.151804 0.088 0.9303
## Recipient4 0.808038 3.027481 0.267 0.7901
## Recipient5 -0.286355 2.374705 -0.121 0.9043
## Recipient6 2.353592 2.306637 1.020 0.3102
## Recipient7 -0.567281 3.130081 -0.181 0.8566
## Recipient8 -5.934832 3.047413 -1.947 0.0545 .
## Recipient9 1.010484 2.610790 0.387 0.6996
## Recipient10 4.078893 3.024122 1.349 0.1806
## Recipient11 -1.404488 2.981273 -0.471 0.6387
## Recipient12 2.791107 2.298540 1.214 0.2277
## Recipient13 -0.547292 2.453803 -0.223 0.8240
## Recipient14 0.526095 2.291091 0.230 0.8189
## Recipient15 0.622225 2.235062 0.278 0.7813
## Recipient16 -0.684972 2.619997 -0.261 0.7943
## Recipient17 -2.472411 2.367061 -1.045 0.2989
## Recipient18 10.154978 35.797790 0.284 0.7773
## Recipient19 -4.306169 2.942373 -1.464 0.1467
## Recipient20 -3.203620 2.496061 -1.283 0.2025
## Treatment1 -1.893005 1.848470 -1.024 0.3084
## Location -0.019891 0.009668 -2.057 0.0424 *
## Recipient1:Treatment1 2.352225 2.404473 0.978 0.3305
## Recipient2:Treatment1 -0.941907 2.596674 -0.363 0.7176
## Recipient3:Treatment1 -3.329951 3.213655 -1.036 0.3028
## Recipient4:Treatment1 -1.084230 3.066302 -0.354 0.7244
## Recipient5:Treatment1 -1.099513 2.453497 -0.448 0.6551
## Recipient6:Treatment1 0.457123 2.351878 0.194 0.8463
## Recipient7:Treatment1 4.337949 3.320822 1.306 0.1946
## Recipient8:Treatment1 -0.112745 3.160043 -0.036 0.9716
## Recipient9:Treatment1 -0.458074 2.711053 -0.169 0.8662
## Recipient10:Treatment1 -0.245309 3.092275 -0.079 0.9369
## Recipient11:Treatment1 4.657866 2.990033 1.558 0.1226
## Recipient12:Treatment1 -0.696892 2.345516 -0.297 0.7670
## Recipient13:Treatment1 1.965994 2.533006 0.776 0.4396
## Recipient14:Treatment1 -0.612205 2.332819 -0.262 0.7936
## Recipient15:Treatment1 1.700431 2.264388 0.751 0.4546
## Recipient16:Treatment1 -2.366163 2.600397 -0.910 0.3652
## Recipient17:Treatment1 -2.727497 2.455897 -1.111 0.2696
## Recipient18:Treatment1 -4.755957 35.782405 -0.133 0.8945
## Recipient19:Treatment1 3.680096 2.990986 1.230 0.2216
## Recipient20:Treatment1 NA NA NA NA
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.357 on 94 degrees of freedom
## (89 observations deleted due to missingness)
## Multiple R-squared: 0.4297, Adjusted R-squared: 0.1809
## F-statistic: 1.727 on 41 and 94 DF, p-value: 0.01557
plot(GxE_Cd_lm0)
## Warning: not plotting observations with leverage one:
## 88, 100, 103, 104, 105, 108, 126
hist(GxE_Cd_lm0$residuals)
#Anova(GxE_Cd_lm0, type = 3)
#Biomass
GxE_BIO_lm1<- lm(log(AG_biomass_mg) ~ Treatment*Recipient + Location, data = BB_Fplants)
summary(GxE_BIO_lm1)
##
## Call:
## lm(formula = log(AG_biomass_mg) ~ Treatment * Recipient + Location,
## data = BB_Fplants)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.8005 -0.3874 0.0000 0.4115 1.5188
##
## Coefficients: (5 not defined because of singularities)
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.963448 0.415111 7.139 1.87e-09 ***
## Treatment1 -0.324380 0.526988 -0.616 0.5407
## Recipient1 1.254641 0.615808 2.037 0.0463 *
## Recipient2 0.122633 0.656787 0.187 0.8525
## Recipient3 0.075662 0.541317 0.140 0.8893
## Recipient4 -0.339050 1.076925 -0.315 0.7540
## Recipient5 -0.094108 0.489908 -0.192 0.8484
## Recipient6 -0.333997 0.487835 -0.685 0.4963
## Recipient7 0.516699 0.596778 0.866 0.3902
## Recipient8 -0.094196 0.633872 -0.149 0.8824
## Recipient9 -0.480734 0.544929 -0.882 0.3814
## Recipient10 0.374070 0.559280 0.669 0.5063
## Recipient11 -0.073496 0.536761 -0.137 0.8916
## Recipient12 0.226105 0.518772 0.436 0.6646
## Recipient13 0.940582 1.074998 0.875 0.3853
## Recipient14 0.634316 0.867155 0.731 0.4675
## Recipient15 -0.416628 0.552705 -0.754 0.4541
## Recipient16 -0.218902 0.497355 -0.440 0.6615
## Recipient17 0.239655 0.560705 0.427 0.6707
## Recipient18 0.064038 1.212456 0.053 0.9581
## Recipient19 -0.601921 0.508993 -1.183 0.2419
## Recipient20 0.571359 0.454923 1.256 0.2143
## Location -0.002966 0.002551 -1.162 0.2499
## Treatment1:Recipient1 -0.377243 0.728877 -0.518 0.6068
## Treatment1:Recipient2 -0.359942 0.753487 -0.478 0.6347
## Treatment1:Recipient3 -0.517830 0.653391 -0.793 0.4313
## Treatment1:Recipient4 NA NA NA NA
## Treatment1:Recipient5 0.136011 0.615624 0.221 0.8259
## Treatment1:Recipient6 0.869847 0.607739 1.431 0.1578
## Treatment1:Recipient7 -0.194714 0.729682 -0.267 0.7905
## Treatment1:Recipient8 0.023857 0.762582 0.031 0.9752
## Treatment1:Recipient9 0.368877 0.660130 0.559 0.5785
## Treatment1:Recipient10 0.036383 0.658607 0.055 0.9561
## Treatment1:Recipient11 0.222124 0.656248 0.338 0.7362
## Treatment1:Recipient12 0.275298 0.617586 0.446 0.6575
## Treatment1:Recipient13 NA NA NA NA
## Treatment1:Recipient14 NA NA NA NA
## Treatment1:Recipient15 -0.375889 0.663842 -0.566 0.5735
## Treatment1:Recipient16 0.621115 0.613801 1.012 0.3159
## Treatment1:Recipient17 0.095373 0.651056 0.146 0.8841
## Treatment1:Recipient18 NA NA NA NA
## Treatment1:Recipient19 0.953002 0.638134 1.493 0.1408
## Treatment1:Recipient20 NA NA NA NA
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.8834 on 57 degrees of freedom
## (130 observations deleted due to missingness)
## Multiple R-squared: 0.3981, Adjusted R-squared: 0.007312
## F-statistic: 1.019 on 37 and 57 DF, p-value: 0.4669
plot(GxE_BIO_lm1)
## Warning: not plotting observations with leverage one:
## 3, 11, 31, 61, 64, 77, 81
hist(GxE_BIO_lm1$residuals)
#Anova(GxE_BIO_lm1, type = 3)
#Est. Fecundity - high VIFs
GxE_FEC_lm1 <- lm(log(total_est_seed_production) ~ Treatment*Recipient, data = filter(BB_Fplants, surv_to_seedprod == 1))
summary(GxE_FEC_lm1)
##
## Call:
## lm(formula = log(total_est_seed_production) ~ Treatment * Recipient,
## data = filter(BB_Fplants, surv_to_seedprod == 1))
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.1809 -0.5280 0.0000 0.5066 2.1462
##
## Coefficients: (8 not defined because of singularities)
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.47398 0.27833 5.296 2.79e-06 ***
## Treatment1 -0.26722 0.40538 -0.659 0.5129
## Recipient1 0.32033 0.43246 0.741 0.4624
## Recipient2 0.04695 0.58611 0.080 0.9365
## Recipient3 0.78701 0.50740 1.551 0.1273
## Recipient4 0.52478 0.57940 0.906 0.3695
## Recipient5 -0.65746 0.92313 -0.712 0.4797
## Recipient6 -0.37846 0.84340 -0.449 0.6556
## Recipient7 -0.14466 0.49963 -0.290 0.7734
## Recipient8 -0.51361 1.12904 -0.455 0.6512
## Recipient9 -0.23153 0.70713 -0.327 0.7447
## Recipient10 0.04753 0.50740 0.094 0.9257
## Recipient11 -0.08768 0.70713 -0.124 0.9018
## Recipient12 1.27273 0.44583 2.855 0.0063 **
## Recipient13 -0.25378 0.46725 -0.543 0.5895
## Recipient14 0.17314 0.57940 0.299 0.7663
## Recipient15 -0.54432 0.51505 -1.057 0.2958
## Recipient16 1.53462 0.58611 2.618 0.0117 *
## Recipient17 0.52657 0.61223 0.860 0.3939
## Recipient18 -0.61498 0.80056 -0.768 0.4461
## Recipient19 0.03569 0.38457 0.093 0.9264
## Treatment1:Recipient1 -0.04178 0.53808 -0.078 0.9384
## Treatment1:Recipient2 0.17871 0.69090 0.259 0.7970
## Treatment1:Recipient3 -0.75131 0.59997 -1.252 0.2164
## Treatment1:Recipient4 NA NA NA NA
## Treatment1:Recipient5 NA NA NA NA
## Treatment1:Recipient6 NA NA NA NA
## Treatment1:Recipient7 0.70066 0.59997 1.168 0.2485
## Treatment1:Recipient8 NA NA NA NA
## Treatment1:Recipient9 0.41106 0.79615 0.516 0.6080
## Treatment1:Recipient10 0.06523 0.59997 0.109 0.9139
## Treatment1:Recipient11 0.26722 0.79615 0.336 0.7386
## Treatment1:Recipient12 0.40326 0.54888 0.735 0.4660
## Treatment1:Recipient13 0.15775 0.56642 0.279 0.7818
## Treatment1:Recipient14 NA NA NA NA
## Treatment1:Recipient15 NA NA NA NA
## Treatment1:Recipient16 -0.23074 0.69090 -0.334 0.7398
## Treatment1:Recipient17 0.05842 0.69090 0.085 0.9330
## Treatment1:Recipient18 NA NA NA NA
## Treatment1:Recipient19 NA NA NA NA
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.969 on 49 degrees of freedom
## Multiple R-squared: 0.4694, Adjusted R-squared: 0.1337
## F-statistic: 1.398 on 31 and 49 DF, p-value: 0.1442
plot(GxE_FEC_lm1 )
## Warning: not plotting observations with leverage one:
## 1, 15, 18, 27, 50, 57, 58, 81
hist(GxE_FEC_lm1 $residuals)
#poor sample sizes...
#Anova(GxE_FEC_lm1, type = 3)
#not filtered version
GxE_FEC_lm1 <- lm(log(total_est_seed_production+1) ~ Treatment*Recipient + Location, data = BB_Fplants)
summary(GxE_FEC_lm1)
##
## Call:
## lm(formula = log(total_est_seed_production + 1) ~ Treatment *
## Recipient + Location, data = BB_Fplants)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.0545 -0.5562 -0.0594 0.5517 3.4110
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.787164 0.154816 5.085 9.11e-07 ***
## Treatment1 -0.190022 0.073031 -2.602 0.01003 *
## Recipient1 0.634778 0.291562 2.177 0.03076 *
## Recipient2 -0.067273 0.293698 -0.229 0.81908
## Recipient3 0.630463 0.305937 2.061 0.04075 *
## Recipient4 0.471419 0.335670 1.404 0.16190
## Recipient5 -0.531484 0.271352 -1.959 0.05168 .
## Recipient6 -0.441480 0.278092 -1.588 0.11413
## Recipient7 0.002969 0.286450 0.010 0.99174
## Recipient8 -0.600392 0.332476 -1.806 0.07260 .
## Recipient9 -0.365955 0.312343 -1.172 0.24287
## Recipient10 0.459717 0.343476 1.338 0.18243
## Recipient11 -0.358700 0.312115 -1.149 0.25196
## Recipient12 0.661798 0.259532 2.550 0.01160 *
## Recipient13 0.256727 0.304442 0.843 0.40018
## Recipient14 -0.177006 0.312286 -0.567 0.57154
## Recipient15 -0.182747 0.271249 -0.674 0.50134
## Recipient16 0.205357 0.286514 0.717 0.47445
## Recipient17 0.294937 0.325984 0.905 0.36679
## Recipient18 -0.725604 0.523038 -1.387 0.16705
## Recipient19 -0.380979 0.332366 -1.146 0.25319
## Recipient20 0.609437 0.344273 1.770 0.07837 .
## Location -0.001179 0.001595 -0.739 0.46072
## Treatment1:Recipient1 -0.172079 0.291540 -0.590 0.55576
## Treatment1:Recipient2 0.405328 0.296118 1.369 0.17275
## Treatment1:Recipient3 -0.552983 0.306535 -1.804 0.07289 .
## Treatment1:Recipient4 -0.970342 0.337457 -2.875 0.00452 **
## Treatment1:Recipient5 0.333668 0.271409 1.229 0.22051
## Treatment1:Recipient6 0.401266 0.277675 1.445 0.15015
## Treatment1:Recipient7 0.339063 0.284643 1.191 0.23513
## Treatment1:Recipient8 0.262794 0.332455 0.790 0.43028
## Treatment1:Recipient9 0.155572 0.312706 0.498 0.61943
## Treatment1:Recipient10 -0.148248 0.340498 -0.435 0.66380
## Treatment1:Recipient11 0.109990 0.311911 0.353 0.72477
## Treatment1:Recipient12 0.153697 0.258168 0.595 0.55236
## Treatment1:Recipient13 -0.137117 0.303372 -0.452 0.65182
## Treatment1:Recipient14 -0.339944 0.311916 -1.090 0.27722
## Treatment1:Recipient15 -0.330496 0.271345 -1.218 0.22480
## Treatment1:Recipient16 0.373900 0.285060 1.312 0.19129
## Treatment1:Recipient17 -0.284723 0.323522 -0.880 0.37998
## Treatment1:Recipient18 0.192616 0.520591 0.370 0.71182
## Treatment1:Recipient19 0.490070 0.332423 1.474 0.14215
## Treatment1:Recipient20 -0.177339 0.341265 -0.520 0.60394
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.9891 on 182 degrees of freedom
## Multiple R-squared: 0.2912, Adjusted R-squared: 0.1276
## F-statistic: 1.78 on 42 and 182 DF, p-value: 0.005161
plot(GxE_FEC_lm1 )
## Warning: not plotting observations with leverage one:
## 145
hist(GxE_FEC_lm1 $residuals)
#Anova(GxE_FEC_lm1, type = 3)